Back to Details

data-analysis-agent-business-intelligence

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Data Analysis Agent — Business Intelligence Skill

Data Analysis Agent — 商业智能技能

Skill by ara.so — AI Agent Skills collection.
Data Analysis Agent is a conversational business intelligence system that enables non-technical users to perform data analysis through natural language. Users can upload Excel/CSV files or connect to databases, then ask questions in plain language. The system automatically understands intent, generates SQL, executes queries, recommends charts, and provides business insights with real-time SSE (Server-Sent Events) streaming output.
由ara.so提供的Skill — AI Agent技能合集。
Data Analysis Agent是一款对话式商业智能系统，可让非技术用户通过自然语言执行数据分析。用户可以上传Excel/CSV文件或连接数据库，然后用通俗语言提出问题。系统会自动理解意图、生成SQL、执行查询、推荐图表，并通过实时SSE（Server-Sent Events）流式输出提供商业洞察。

What It Does

功能特性

  • Natural Language Queries: Ask questions in plain English instead of writing SQL
  • Auto SQL Generation: Converts natural language to optimized SQL queries
  • Smart Visualization: Automatically recommends from 43+ chart types based on data patterns
  • Multi-Source Support: Excel, CSV, SQLite, MySQL, PostgreSQL, SQL Server, Google Sheets
  • Streaming Analysis: Real-time SSE output showing analysis progress
  • Advanced Analytics: K-Means clustering, decision trees, outlier handling, decile analysis
  • Report Generation: Export to Excel, Word, PowerPoint
  • MCP Integration: Extend capabilities with Model Context Protocol tools
  • Knowledge Base: Upload business documents to improve domain understanding
  • 自然语言查询：用通俗英语提问，无需编写SQL
  • 自动SQL生成：将自然语言转换为优化后的SQL查询语句
  • 智能可视化：根据数据模式从43+种图表类型中自动推荐
  • 多数据源支持：Excel、CSV、SQLite、MySQL、PostgreSQL、SQL Server、Google Sheets
  • 流式分析：实时SSE输出展示分析进度
  • 高级分析：K-Means聚类、决策树、异常值处理、十分位分析
  • 报告生成：导出至Excel、Word、PowerPoint
  • MCP集成：通过Model Context Protocol工具扩展功能
  • 知识库：上传业务文档提升领域理解能力

Installation

安装方法

Method 1: Download Release Package (Recommended)

方法1：下载发布包（推荐）

bash
undefined
bash
undefined

Download from releases page

从发布页面下载

Extract the archive

解压压缩包

Navigate to directory

进入目录

Windows

Windows系统

start.bat
start.bat

macOS (first time - grant permissions)

macOS系统（首次使用需授权）

chmod +x start.command ./start.command
chmod +x start.command ./start.command

Access at http://localhost:5001

访问地址：http://localhost:5001

undefined
undefined

Method 2: Clone from GitHub

方法2：从GitHub克隆

bash
undefined
bash
undefined

Clone repository

克隆仓库

git clone https://github.com/Zafer-Liu/Data-Analysis-Agent.git cd Data-Analysis-Agent
git clone https://github.com/Zafer-Liu/Data-Analysis-Agent.git cd Data-Analysis-Agent

Install dependencies

安装依赖

pip install -r requirements.txt
pip install -r requirements.txt

Start server

启动服务器

python app.py
python app.py

Access at http://localhost:5001

访问地址：http://localhost:5001

undefined
undefined

Method 3: One-Line Install

方法3：一键安装

Windows (PowerShell):
powershell
iwr -useb https://raw.githubusercontent.com/Zafer-Liu/Data-Analysis-Agent/main/install.ps1 | iex
macOS/Linux:
bash
curl -fsSL https://raw.githubusercontent.com/Zafer-Liu/Data-Analysis-Agent/main/install.sh | sh
data-analysis-agent
Windows（PowerShell）：
powershell
iwr -useb https://raw.githubusercontent.com/Zafer-Liu/Data-Analysis-Agent/main/install.ps1 | iex
macOS/Linux：
bash
curl -fsSL https://raw.githubusercontent.com/Zafer-Liu/Data-Analysis-Agent/main/install.sh | sh
data-analysis-agent

Configuration

配置说明

LLM Setup

LLM设置

Configure your LLM provider in the web UI sidebar (⚙ icon):
python
undefined
在Web UI侧边栏（⚙图标）中配置LLM提供商：
python
undefined

Supported providers

支持的提供商

providers = { "deepseek": { "base_url": "https://api.deepseek.com", "model": "deepseek-chat", "api_key": os.getenv("DEEPSEEK_API_KEY") }, "openai": { "base_url": "https://api.openai.com/v1", "model": "gpt-4o-mini", "api_key": os.getenv("OPENAI_API_KEY") }, "anthropic": { "base_url": "https://api.anthropic.com", "model": "claude-3-5-haiku-20241022", "api_key": os.getenv("ANTHROPIC_API_KEY") } }
undefined
providers = { "deepseek": { "base_url": "https://api.deepseek.com", "model": "deepseek-chat", "api_key": os.getenv("DEEPSEEK_API_KEY") }, "openai": { "base_url": "https://api.openai.com/v1", "model": "gpt-4o-mini", "api_key": os.getenv("OPENAI_API_KEY") }, "anthropic": { "base_url": "https://api.anthropic.com", "model": "claude-3-5-haiku-20241022", "api_key": os.getenv("ANTHROPIC_API_KEY") } }
undefined

Database Connection

数据库连接

MySQL/PostgreSQL:
python
undefined
MySQL/PostgreSQL：
python
undefined

Connection string format

连接字符串格式

mysql_connection = "mysql+pymysql://{username}:{password}@{host}:{port}/{database}" postgres_connection = "postgresql://{username}:{password}@{host}:{port}/{database}"
mysql_connection = "mysql+pymysql://{username}:{password}@{host}:{port}/{database}" postgres_connection = "postgresql://{username}:{password}@{host}:{port}/{database}"

Example (use environment variables)

示例（使用环境变量）

import os db_url = f"mysql+pymysql://{os.getenv('DB_USER')}:{os.getenv('DB_PASS')}@localhost:3306/sales_db"

**Google Sheets:**
```python
import os db_url = f"mysql+pymysql://{os.getenv('DB_USER')}:{os.getenv('DB_PASS')}@localhost:3306/sales_db"

**Google Sheets：**
```python

Configure Google Sheets API credentials

配置Google Sheets API凭证

Upload credentials JSON in the web UI

在Web UI中上传凭证JSON文件

Enter Sheet ID to connect

输入Sheet ID进行连接

undefined
undefined

Slash Commands

斜杠命令

Core commands for specialized analysis:
bash
/chart        # Force chart generation priority
/sql          # Execute SQL directly
/analyze      # Deep statistical analysis
/tree         # Decision tree modeling
/kmeans       # K-Means clustering
/data         # Data exploration and preview
/inset        # Missing value imputation
/winsorize    # Winsorize outliers (replace extremes)
/trimming     # Trim outliers (remove extremes)
/export       # Export data file
/report       # Generate Word/PDF report
/ppt          # Generate PowerPoint presentation
/status       # Check task status
用于专项分析的核心命令：
bash
/chart        # 优先强制生成图表
/sql          # 直接执行SQL语句
/analyze      # 深度统计分析
/tree         # 决策树建模
/kmeans       # K-Means聚类
/data         # 数据探索与预览
/inset        # 缺失值插补
/winsorize    # 缩尾处理异常值（替换极端值）
/trimming     # 截尾处理异常值（移除极端值）
/export       # 导出数据文件
/report       # 生成Word/PDF报告
/ppt          # 生成PowerPoint演示文稿
/status       # 检查任务状态

Code Examples

代码示例

Natural Language Query

自然语言查询

python
undefined
python
undefined

Example query flow in the web UI

Web UI中的示例查询流程

user_query = "What is the sales trend for the last 12 months?"
user_query = "过去12个月的销售趋势如何？"

System automatically:

系统自动执行以下操作：

1. Analyzes schema

1. 分析数据结构

2. Generates SQL:

2. 生成SQL语句：

SELECT DATE_FORMAT(order_date, '%Y-%m') as month,

SELECT DATE_FORMAT(order_date, '%Y-%m') as month,

SUM(sales_amount) as total_sales

SUM(sales_amount) as total_sales

FROM sales

FROM sales

WHERE order_date >= DATE_SUB(NOW(), INTERVAL 12 MONTH)

WHERE order_date >= DATE_SUB(NOW(), INTERVAL 12 MONTH)

GROUP BY month

GROUP BY month

ORDER BY month

ORDER BY month

3. Executes query

3. 执行查询

4. Recommends Line_Chart visualization

4. 推荐Line_Chart可视化图表

5. Generates insights

5. 生成洞察结论

undefined
undefined

Programmatic API Usage

程序化API使用

python
from flask import Flask
from modules.agent_core import analyze_query
import os
python
from flask import Flask
from modules.agent_core import analyze_query
import os

Initialize

初始化

app = Flask(name)
app = Flask(name)

Configure LLM

配置LLM

llm_config = { "provider": "deepseek", "api_key": os.getenv("DEEPSEEK_API_KEY"), "base_url": "https://api.deepseek.com", "model": "deepseek-chat" }
llm_config = { "provider": "deepseek", "api_key": os.getenv("DEEPSEEK_API_KEY"), "base_url": "https://api.deepseek.com", "model": "deepseek-chat" }

Analyze data

分析数据

def query_data(question, data_source): """ Question: Natural language query data_source: Path to CSV/Excel or database connection string """ result = analyze_query( question=question, data_source=data_source, llm_config=llm_config, stream=True # Enable SSE streaming ) return result
def query_data(question, data_source): """ Question: 自然语言查询问题 data_source: CSV/Excel文件路径或数据库连接字符串 """ result = analyze_query( question=question, data_source=data_source, llm_config=llm_config, stream=True # 启用SSE流式输出 ) return result

Example usage

使用示例

response = query_data( question="Which region has the highest profit?", data_source="./data/sales_data.csv" )
undefined
response = query_data( question="哪个地区的利润最高？", data_source="./data/sales_data.csv" )
undefined

Chart Generation

图表生成

python
undefined
python
undefined

Force chart generation with /chart command

使用/chart命令强制生成图表

user_input = "/chart user growth over time"
user_input = "/chart 用户增长趋势"

System response includes:

系统响应包含：

- SQL query

- SQL查询语句

- Data result

- 数据结果

- Chart recommendation (e.g., Line_Chart, Area_Chart)

- 图表推荐（如Line_Chart、Area_Chart）

- Interactive Plotly visualization

- 交互式Plotly可视化图表

- Chart saved to ./outputs/charts/

- 图表保存至./outputs/charts/

undefined
undefined

Advanced Analytics: K-Means Clustering

高级分析：K-Means聚类

python
undefined
python
undefined

Use /kmeans command

使用/kmeans命令

query = "/kmeans segment customers by purchase behavior"
query = "/kmeans 根据购买行为细分客户"

System performs:

系统执行以下操作：

1. Feature selection

1. 特征选择

2. Data normalization

2. 数据标准化

3. Optimal cluster determination (elbow method)

3. 确定最优聚类数（肘部法）

4. K-Means clustering

4. K-Means聚类

5. Cluster visualization

5. 聚类可视化

6. Business interpretation of segments

6. 细分群体的商业解读

undefined
undefined

MCP Tool Integration

MCP工具集成

python
undefined
python
undefined

Enable MCP tools in configuration

在配置中启用MCP工具

mcp_config = { "enabled": True, "tools": [ { "name": "calculator", "endpoint": "http://localhost:8000/mcp/calculator" }, { "name": "code_executor", "endpoint": "http://localhost:8000/mcp/execute" } ] }
mcp_config = { "enabled": True, "tools": [ { "name": "calculator", "endpoint": "http://localhost:8000/mcp/calculator" }, { "name": "code_executor", "endpoint": "http://localhost:8000/mcp/execute" } ] }

Agent automatically invokes MCP tools when needed

Agent会在需要时自动调用MCP工具

Example: Complex financial calculations beyond SQL

示例：超出SQL能力范围的复杂财务计算

undefined
undefined

Knowledge Base Integration

知识库集成

python
undefined
python
undefined

Upload business documents via web UI

通过Web UI上传业务文档

Supported formats: .docx, .xlsx, .pdf, .txt

支持格式：.docx, .xlsx, .pdf, .txt

Documents are vectorized and stored

文档会被向量化并存储

Agent retrieves relevant context automatically

Agent会自动检索相关上下文

Example query with knowledge enhancement:

带知识增强的示例查询：

query = "Analyze Q4 sales using last year's strategic priorities"
query = "结合去年的战略重点分析第四季度销售额"

Agent retrieves relevant strategic docs from knowledge base

Agent从知识库中检索相关战略文档

Provides context-aware insights aligned with business goals

提供符合业务目标的上下文感知洞察

undefined
undefined

Common Patterns

常见分析模式

Pattern 1: Time Series Analysis

模式1：时间序列分析

python
undefined
python
undefined

Natural language input

自然语言输入

"Show me monthly revenue trend with year-over-year comparison"
"展示月度收入趋势及同比对比"

Generated SQL pattern

生成的SQL模式

""" SELECT DATE_FORMAT(order_date, '%Y-%m') as month, SUM(revenue) as current_revenue, LAG(SUM(revenue), 12) OVER (ORDER BY DATE_FORMAT(order_date, '%Y-%m')) as prev_year_revenue, ((SUM(revenue) - LAG(SUM(revenue), 12) OVER (ORDER BY DATE_FORMAT(order_date, '%Y-%m'))) / LAG(SUM(revenue), 12) OVER (ORDER BY DATE_FORMAT(order_date, '%Y-%m'))) * 100 as yoy_growth FROM orders GROUP BY month ORDER BY month """
""" SELECT DATE_FORMAT(order_date, '%Y-%m') as month, SUM(revenue) as current_revenue, LAG(SUM(revenue), 12) OVER (ORDER BY DATE_FORMAT(order_date, '%Y-%m')) as prev_year_revenue, ((SUM(revenue) - LAG(SUM(revenue), 12) OVER (ORDER BY DATE_FORMAT(order_date, '%Y-%m'))) / LAG(SUM(revenue), 12) OVER (ORDER BY DATE_FORMAT(order_date, '%Y-%m'))) * 100 as yoy_growth FROM orders GROUP BY month ORDER BY month """

Auto-recommended chart: Line_Chart with dual axis

自动推荐图表：双轴Line_Chart

undefined
undefined

Pattern 2: Outlier Detection & Handling

模式2：异常值检测与处理

python
undefined
python
undefined

Detect outliers

检测异常值

query = "Find outliers in customer spending data"
query = "找出客户消费数据中的异常值"

Apply winsorization

应用缩尾处理

query = "/winsorize cap extreme values at 5th and 95th percentile"
query = "/winsorize 将极端值限制在5%和95%分位数"

Or trim outliers

或应用截尾处理

query = "/trimming remove values beyond 3 standard deviations"
undefined
query = "/trimming 移除超出3倍标准差的值"
undefined

Pattern 3: Cohort Analysis

模式3：同期群分析

python
query = "Create cohort analysis for user retention by signup month"
python
query = "按注册月份创建用户留存同期群分析"

System generates cohort table and retention curve

系统生成同期群表格和留存曲线

Automatically suggests Heatmap for cohort visualization

自动推荐Heatmap用于同期群可视化

undefined
undefined

Pattern 4: Export Workflow

模式4：导出工作流

python
undefined
python
undefined

After analysis, export results

分析完成后导出结果

"/export save cleaned data to Excel"
"/export 将清洗后的数据保存至Excel"

Generate comprehensive report

生成综合报告

"/report create analysis report with all charts and insights"
"/report 创建包含所有图表和洞察的分析报告"

Create presentation

创建演示文稿

"/ppt generate executive summary presentation"
"/ppt 生成高管摘要演示文稿"

Files saved to ./outputs/ directory

文件保存至./outputs/目录

undefined
undefined

Chart Types Reference

图表类型参考

The system auto-selects from 43 chart types:
python
chart_categories = {
    "COMPARING": [
        "Bar_Chart", "Grouped_Bar_Chart", "Stacked_Bar_Chart",
        "Diverging_Bar_Chart", "Marimekko_ABS", "Waterfall", "Sankey_Chart"
    ],
    "TIME": [
        "Line_Chart", "Area_Chart", "Stacked_Area_Chart",
        "Slope_Chart", "Bump_Chart", "Sparkline"
    ],
    "DISTRIBUTION": [
        "Histogram_Pareto_chart", "Box-and-Whisker_Plot", 
        "Violin_Chart", "Ridgeline_Plot", "Beeswarm_Plot"
    ],
    "GEOSPATIAL": [
        "Choropleth_Map", "Dot_Density_Map", "Flow_Map"
    ],
    "RELATIONSHIP": [
        "Scatter_Plot", "Bubble_Plot", "Network_Diagram",
        "Chord_Diagram", "Parallel_Coordinates_Plot"
    ],
    "PART-TO-WHOLE": [
        "Pie_Chart", "Treemap", "Sunburst_Diagram", "Nightingale_Chart"
    ]
}
系统会从43种图表类型中自动选择：
python
chart_categories = {
    "COMPARING": [
        "Bar_Chart", "Grouped_Bar_Chart", "Stacked_Bar_Chart",
        "Diverging_Bar_Chart", "Marimekko_ABS", "Waterfall", "Sankey_Chart"
    ],
    "TIME": [
        "Line_Chart", "Area_Chart", "Stacked_Area_Chart",
        "Slope_Chart", "Bump_Chart", "Sparkline"
    ],
    "DISTRIBUTION": [
        "Histogram_Pareto_chart", "Box-and-Whisker_Plot", 
        "Violin_Chart", "Ridgeline_Plot", "Beeswarm_Plot"
    ],
    "GEOSPATIAL": [
        "Choropleth_Map", "Dot_Density_Map", "Flow_Map"
    ],
    "RELATIONSHIP": [
        "Scatter_Plot", "Bubble_Plot", "Network_Diagram",
        "Chord_Diagram", "Parallel_Coordinates_Plot"
    ],
    "PART-TO-WHOLE": [
        "Pie_Chart", "Treemap", "Sunburst_Diagram", "Nightingale_Chart"
    ]
}

Troubleshooting

故障排除

Issue: "LLM Not Configured"

问题："LLM未配置"

Solution:
python
undefined
解决方案：
python
undefined

Set API key in web UI sidebar or environment

在Web UI侧边栏或环境变量中设置API密钥

export DEEPSEEK_API_KEY="your-key-here"
export DEEPSEEK_API_KEY="your-key-here"

Or configure in the settings panel (⚙ icon)

或在设置面板（⚙图标）中配置

undefined
undefined

Issue: Database Connection Failed

问题：数据库连接失败

Solution:
python
undefined
解决方案：
python
undefined

Verify connection string format

验证连接字符串格式

MySQL example:

MySQL示例：

connection_string = "mysql+pymysql://user:password@host:port/database"
connection_string = "mysql+pymysql://user:password@host:port/database"

Test connection

测试连接

import pymysql conn = pymysql.connect( host=os.getenv('DB_HOST'), user=os.getenv('DB_USER'), password=os.getenv('DB_PASS'), database=os.getenv('DB_NAME') ) conn.close()
undefined
import pymysql conn = pymysql.connect( host=os.getenv('DB_HOST'), user=os.getenv('DB_USER'), password=os.getenv('DB_PASS'), database=os.getenv('DB_NAME') ) conn.close()
undefined

Issue: Charts Not Displaying

问题：图表不显示

Solution:
bash
undefined
解决方案：
bash
undefined

Charts are saved locally in:

图表本地保存路径：

./outputs/charts/
./outputs/charts/

Check browser console for errors

检查浏览器控制台错误

Verify Plotly.js is loaded

验证Plotly.js是否加载

Clear browser cache if needed

必要时清除浏览器缓存

undefined
undefined

Issue: Slow Query Performance

问题：查询性能缓慢

Solution:
python
undefined
解决方案：
python
undefined

Add database indexes on frequently queried columns

在频繁查询的列上添加数据库索引

Example for MySQL:

MySQL示例：

""" CREATE INDEX idx_order_date ON orders(order_date); CREATE INDEX idx_customer_id ON orders(customer_id); """
""" CREATE INDEX idx_order_date ON orders(order_date); CREATE INDEX idx_customer_id ON orders(customer_id); """

Or use /sql command to optimize query manually

或使用/sql命令手动优化查询

undefined
undefined

Issue: Incorrect SQL Generation

问题：SQL生成错误

Solution:
python
undefined
解决方案：
python
undefined

Provide more context in natural language query

在自然语言查询中提供更多上下文

Bad: "sales by region"

不佳示例："按地区统计销售额"

Good: "total sales amount grouped by region for 2024"

良好示例："2024年按地区统计的总销售额"

Or use /sql command to write SQL directly

或使用/sql命令直接编写SQL

/sql SELECT region, SUM(amount) FROM sales WHERE YEAR(date) = 2024 GROUP BY region
undefined
/sql SELECT region, SUM(amount) FROM sales WHERE YEAR(date) = 2024 GROUP BY region
undefined

Issue: Memory Error on Large Files

问题：大文件内存错误

Solution:
python
undefined
解决方案：
python
undefined

For large CSVs, use chunked reading

处理大CSV文件时使用分块读取

import pandas as pd chunks = pd.read_csv('large_file.csv', chunksize=10000) for chunk in chunks: # Process each chunk pass
import pandas as pd chunks = pd.read_csv('large_file.csv', chunksize=10000) for chunk in chunks: # 处理每个分块 pass

Or import to SQLite/database first, then query

或先导入SQLite/数据库，再进行查询

undefined
undefined

Advanced Usage

高级用法

Custom Chart Configuration

自定义图表配置

python
undefined
python
undefined

Modify chart templates in modules/chart_generator.py

修改modules/chart_generator.py中的图表模板

def create_custom_chart(data, chart_type): import plotly.graph_objects as go
fig = go.Figure()
# Custom Plotly configuration
fig.update_layout(
    template="plotly_white",
    title_font_size=20,
    # Add custom styling
)
return fig
undefined
def create_custom_chart(data, chart_type): import plotly.graph_objects as go
fig = go.Figure()
# 自定义Plotly配置
fig.update_layout(
    template="plotly_white",
    title_font_size=20,
    # 添加自定义样式
)
return fig
undefined

Extend Analysis Functions

扩展分析功能

python
undefined
python
undefined

Add custom analysis in modules/analyzer.py

在modules/analyzer.py中添加自定义分析函数

def custom_analysis(df, params): """ Custom statistical analysis """ from scipy import stats
result = stats.ttest_ind(
    df[params['group1']], 
    df[params['group2']]
)

return {
    "statistic": result.statistic,
    "p_value": result.pvalue,
    "interpretation": "Significant" if result.pvalue < 0.05 else "Not significant"
}
undefined
def custom_analysis(df, params): """ 自定义统计分析 """ from scipy import stats
result = stats.ttest_ind(
    df[params['group1']], 
    df[params['group2']]
)

return {
    "statistic": result.statistic,
    "p_value": result.pvalue,
    "interpretation": "显著" if result.pvalue < 0.05 else "不显著"
}
undefined

Streaming Response Handler

流式响应处理

python
undefined
python
undefined

Handle SSE stream in custom integration

在自定义集成中处理SSE流

import requests
def stream_analysis(query): response = requests.get( 'http://localhost:5001/api/analyze', params={'query': query}, stream=True )
for line in response.iter_lines():
    if line:
        decoded = line.decode('utf-8')
        if decoded.startswith('data: '):
            data = decoded[6:]  # Remove 'data: ' prefix
            print(data)  # Process each streaming chunk
undefined
import requests
def stream_analysis(query): response = requests.get( 'http://localhost:5001/api/analyze', params={'query': query}, stream=True )
for line in response.iter_lines():
    if line:
        decoded = line.decode('utf-8')
        if decoded.startswith('data: '):
            data = decoded[6:]  # 移除'data: '前缀
            print(data)  # 处理每个流式数据块
undefined

Environment Variables Reference

环境变量参考

bash
undefined
bash
undefined

LLM Configuration

LLM配置

DEEPSEEK_API_KEY=your_deepseek_key OPENAI_API_KEY=your_openai_key ANTHROPIC_API_KEY=your_anthropic_key
DEEPSEEK_API_KEY=your_deepseek_key OPENAI_API_KEY=your_openai_key ANTHROPIC_API_KEY=your_anthropic_key

Database Credentials

数据库凭证

DB_HOST=localhost DB_PORT=3306 DB_USER=your_username DB_PASS=your_password DB_NAME=your_database
DB_HOST=localhost DB_PORT=3306 DB_USER=your_username DB_PASS=your_password DB_NAME=your_database

Google Sheets (if using)

Google Sheets（如需使用）

GOOGLE_SHEETS_CREDENTIALS_PATH=/path/to/credentials.json
GOOGLE_SHEETS_CREDENTIALS_PATH=/path/to/credentials.json

Server Configuration

服务器配置

FLASK_PORT=5001 FLASK_ENV=production
undefined
FLASK_PORT=5001 FLASK_ENV=production
undefined

Resources

资源链接

  • Repository: https://github.com/Zafer-Liu/Data-Analysis-Agent
  • Documentation: See README.md and Information/ directory
  • MCP Tutorial: Information/MCP_tutorial.md
  • Knowledge Base Guide: Information/repository_tutorial.md
  • Version History: Information/Version_Update_Log.md
  • 仓库地址：https://github.com/Zafer-Liu/Data-Analysis-Agent
  • 文档：查看README.md和Information/目录
  • MCP教程：Information/MCP_tutorial.md
  • 知识库指南：Information/repository_tutorial.md
  • 版本历史：Information/Version_Update_Log.md